1: \begin{abstract}

2: This paper focuses on the estimation of sensitivity matrices in electrical transmission systems. By leveraging a low-rank approximation of certain classes of sensitivity matrices, the paper proposes a robust nuclear norm minimization method to estimate sensitivities from measurements. Relative to existing methods based on the least-squares approach, the proposed method can obtain meaningful estimates with a smaller number of measurements and when the regression model is underdetermined; the method can also identify faulty measurements and handle missing data. Furthermore, an online proximal-gradient method is proposed to estimate sensitivities on-the-fly from real-time measurements; convergence results in terms of dynamic regret are offered in this case. While the proposed approach can be leveraged to estimate various sensitivity coefficients in a power grid, this paper considers power transfer distribution factors as an illustrative example. Numerical tests corroborate the effectiveness of the novel approach.

3: \end{abstract}

4: